PHD Still getting star trails !!

[decoyp]

Hi.

I've just finished building a ROR observatory and have had to reset my mount on the pier because I had to raise the whole thing by about 10 inches (long story).

I've done a re-alignment by drift aligning which I think is OK. There was pretty much no movement after about 10 mins. I'm also using PHD to guide. However I have taken a few test shots wide field just to check things are OK and I've noticed trailing on all the shots.

The PHD graph is also not very flat (no screen grabs I'm afraid). The RA (blue trace) seems to oscillate either side of the horizontal graph line by one whole division up or down from the middle horizontal line. This oscillation is in phase with each exposure taken from the QHY5 camera i.e about every 2 seconds.

DEC trace seems to come into the graph from the top left hand corner and then steadily descend to the horizontal axis and then flatten out.

Can anyone shed any light on this? should I be adjusting some of the brain settings or biasing the counterweights somehow ?

Any help greatly appreciated. Why does this always happen when we get a plague of clear skies !!!!

Thanks

Richard

[Mav359]

Mine has started to do this, both DEC & RA show very small steady amounts of movement off the horizontal line then all of a sudden the DEC does a right turn bottoms out and you don't see it for 20mins the back it comes loiters around the central horizontal line then suddenly off it goes again.

I had my guiding up to 8mins without any star trailing, then this developed but i don't know what I've changed (I'm sure it something I've done!!!). I'm considering striping the lot down and reassembling from scratch. I'm leaning towards balancing issues at this point as I've had this working using only the default PHD settings.

If anyone does have a definitive answer I'd like to know too...

[shaunster]

What guide scope are you using and how many calibration steps did PHD take?

How is your setup balanced?

[Merlin66]

Need more info on the setup, but my bet would be on the balancing.

[Mav359]

What guide scope are you using and how many calibration steps did PHD take?

How is your setup balanced?

I'm using a 9x50 finder scope with a QHY5.

I'm balanced as per the instructions and am confident in RA that it is balanced correctly, it doesn't roll away with the clutch unlocked. As for DEC i'm not so sure, i could get it so that it wouldn't roll away to a point.

The thing i find strange is that this has been working and then something has changed.

[Merlin66]

The cold weather has that effect...even I don't work as well ;-)

[Komet]

May I suggest you try the demo version of Astroart5. Get the guiding going in that and see if the same thing happens. My bet is that it won't.

DEC wandering off seems to be a less than rare issue - certainly happened to me. In Maxim and Astroart - without changing anything else - guiding was spot on.

If the same thing occurs then that would suggest a hardware fault.

[decoyp]

Sorry, I've been at work and haven't had chance to reply to any posts.

I was wondering whether to try another guiding program too. Might do that soon just to see what happens. I will also check the balancing to see if that has an effect. I must admit though I've been reading about Alignmaster and EQmod and am tempted to bite the bullet and go down this route. At least I'll be certain that my polar alignment is spot on - that would take one variable out of the equation.

I know it doesnt help, but I'm glad I,m not the only one having problems like this.

[decoyp]

Right. I had another go at this last night and tried tweaking some values in PHD to get the graph a bit flatter. Still not great but it has improved, however I have noticed something else and have been thinking about my setup. TRhe followwing questions have occured to me ......

1- My test shots show star trailing but only in the leftmost part of the frame. The stars in the right hand side of the frame appear to be significantly better shaped.

2- I am using a DSLR with a 135mm prime focus tele. My guide scope however is a QHY5 with a 400mm tele lense attached. Quite a big difference.

3- Is the difference I am seeing in star shape across the frame due to field rotation because my polar alignment isnt as good as I'd hope?

4- Could the above in point 3 be partly due to the difference in ratio of the scopes and the fact that I can't be certain the guide scope is looking at exactly the same piece of sky as the camera?

Any help much appreciated.

Thanks

Richard.

[Merlin66]

Well...

1. This infers possible field rotation....

2. This set up shoul give good guiding -arc sec/pixel

3. Probably

4.No, the focal lengths, but if the polar alignment is sus, and you're guiding on a star far away from the cente of the imaging FOV, the rotation will be greater. (Usually symmetrical to the guide centre)

[RobH]

I don't entirely agree Ken.

1- My test shots show star trailing but only in the leftmost part of the frame. The stars in the right hand side of the frame appear to be significantly better shaped.

Trailing would show on all of the stars if it was a tracking error....it sounds more like your camera isn't exactly at 90 degrees to the optical axis.

2- I am using a DSLR with a 135mm prime focus tele. My guide scope however is a QHY5 with a 400mm tele lense attached. Quite a big difference.

Shouldn't ba a problem

3- Is the difference I am seeing in star shape across the frame due to field rotation because my polar alignment isnt as good as I'd hope?

Field rotation manifests as trailing around the centre of the FOV. If it's not doing that it's not field rotation.

4- Could the above in point 3 be partly due to the difference in ratio of the scopes and the fact that I can't be certain the guide scope is looking at exactly the same piece of sky as the camera?

The guide scope doesn't need to point at exactly the same point in the sky....near enough will do. If it's a long way off you'll get filed rotation. I never use guide rings so my guidescope never points to exactly the same place, and I've never suffered from any bad effects because of that.

[Jessun]

4.No, the focal lengths, but if the polar alignment is sus, and you're guiding on a star far away from the cente of the imaging FOV, the rotation will be greater. (Usually symmetrical to the guide centre)

Could it be that you guided on a star towards the edge of your imaging scope FOV or even outside of its FOV? How well 'aligned' are your scopes? Could you have cone error?

PHD can not really be blamed as all it does is keep one star still in the guide scope. It has very little idea about where the imaging scope is pointing. True it calibrates, but does not produce a hint if it discovers anything out of the ordinary.

For wide field photography I would assume that you risk errors towards the edges more than for narrow field if any alignment errors are present. Just throwing in the thought.

I am new to all this though.

[decoyp]

Could it be that you guided on a star towards the edge of your imaging scope FOV or even outside of its FOV? How well 'aligned' are your scopes? Could you have cone error?

I think that the FOV of the guide scope could be on the edge of the imaging FOV. It was just a rough test and I foolishly did not consider whether this would be an issue. I suppose it could be cone error I'd know better if I actually knew what cone error was :-) Seriously though I can guess what it is from what's being said here. The two scopes are not particularly aligned - both are mounted on an aluminium plate and I have purposely made them so I can move the guide scope to align to a guide star if necessary. How aligned do they have to be ?? I'm probably within a few degrees (I mean they are not 90 degrees to each other or anything daft like that) .

[Jessun]

I have seen some calculator for this, but can not find it right now. If your polar alignment is perfect you can guide on anything. If the polar alignment isn't, there is a clever formula for how much error is acceptable before field rotation kicks in as a function of PA inaccuracy.

The wider the field the more the edge stars will move corresponding to any errors. (At least that's my understanding). Until you get PA spot on then guiding on or right up close to the imaging target would at least spread the error evenly across the image leaving the central part good and all corners equally 'bad'. When the PA is perfect it doesn't matter.

Maybe you just knocked something during repositioning, if it indeed performs worse than it used to?

[decoyp]

I'll try and re-align both lenses to be pointing at the same thing initially. Probably do this on something unmistakable like Jupiter just so I know both are looking at the same thing. Then I'll try again. I'm sure that as you say PA is an issue and I am going to go down the EQmod route ASAP and use Alignmaster which I've seen the youtube video of and it looks great.

I've heard of people using PHD to drift align but I tried it and just got confused about which way to alter the mount, and ended up with a pretty trashy PA. Anyway I dont't get that much time at the scope and we don't have enough clear nights to waste messing around with PA so I want a nice easy solution.

Thanks for all your help.

BTW what exactly is "Cone Error"

Cheers

Richard

[Merlin66]

Cone error is then the axis of the OTA is misaligned to the RA axis.

ie the OTA is not sitting exactly at 90 degrees to the dec axis.

When polar aligning you use either the polar scope in the mount or the telescope itself..if they're not in tune and parallel to each other you'll still get an error.

( Ever wondered what those two little M5 socket head screws under each end of the Skywatcher dovetails were for?? These are used to correct the cone error by slightly changing the position of the tube rigs and hence the axis of the scope.)

The Skywatcher manual for the EQ mounts has a reason section on how to......

HTH